Allogeneic bone marrow or hematopoietic stem cell transplantation (allo BMT) is an important therapy with curative potential for a variety of malignant and non-malignant diseases. Despite standard prophylactic regimens, graft-versus-host disease (GVHD) continues to limit the success of outcomes in allo BMT patients. Gastrointestinal (GI) GVHD is the predominant contributor to acute GVHD-related mortality. In the first funding period we hypothesized that T cell trafficking to the intestines represented a therapeutic target that could significantly reduce intestinal and systemic GVHD while sparing graft-versus-tumor and improve overall outcome. Among other results we identified the ?47 integrin as an excellent target for the prevention of GI GVHD and we are currently developing a clinical protocol with a neutralizing antibody. In this renewal application, we propose to continue exploring the pathophysiology of intestinal GVHD and focus on the role of the intestinal flora. Early studies in mice and in patients suggested a link between an individual's intestinal microbial flora and his/her propensity for GVHD, but the translation to the clinic (gut decontamination and/or laminar airflow protective environment) remains controversial. The availability of a novel technology, 16s ribosomal RNA (rRNA) gene sequencing, to analyze the composition of the gut flora in much greater detail has resulted in better understanding of the relationship between gut flora and its interaction with intestinal immunity and intestinal epithelil homeostasis. Our preliminary data in mouse models indicate that GVHD is associated with specific changes in the intestinal flora, including a loss of microbial diversity, dysbiosis (imbalance of intestinal flora) and the expansion of Lactobacillus or Enterobacteriaceae, which have opposite effects on GVHD. Moreover, pre-BMT antibiotic use could aggravate GVHD. Our analysis of 66 allo BMT patients shows a) a remarkable loss of microbial diversity in allo-HSCT recipients, which is associated with bacteremia and intestinal domination by Enterococci, Streptococci and Staphylococci, and b) an association between dysbiosis and GVHD. Therefore, we hypothesize that the interactions between intestinal microbiota, allogeneic donor cells, intestinal immunity and intestinal epithelial homeostasis can impact GVHD, and this can be exploited therapeutically to reduce GVHD. We propose in Aim 1 to fully characterize the intestinal microbiota during allo BMT. We will assess the effects of the conditioning and antibiotic regimens and further characterize the GVHD-associated changes (loss of microbial diversity, dysbiosis and expansion of Lactobacillus or Enterobacteriaceae). These experiments are required to define the variables that affect the intestinal flora in allo BMT recipients and provide the necessary background for the experiments in Aim 2, in which we hypothesize that manipulation of the intestinal flora and nutrition can ameliorate intestinal GVHD. We will study the effects of a) introduction of Lactobacillus on GVHD and potential mechanisms of action (suppression of Enterococcus and ??T cells), b) other potentially anti-inflammatory bacteria on GVHD (prioritizing Bifidobacterium), and c) the diet content of aryl hydrocarbon receptor ligands on the intestinal flora and GVHD. In addition, we will explore effects of murine and human intestinal flora on GVHD in germ-free and gnotobiotic mice. The investigators of this proposal are performing similar sequential analyses of the gut flora in allo BMT patients, which fosters a continuing interaction and translatability of the results. Therefore, we expect that these preclinical studies have a high likelihood to improve prophylaxis and treatment of intestinal GVHD as well as overall outcome in allo BMT patients.